Back to HiDAC: Administering the Optimal Treatment for Children With Down Syndrome and Acute Myeloid Leukemia

Children with Down syndrome (DS) have an increased risk of multiple serious health conditions, and they have a particularly high risk of developing acute leukemia. The estimated risk is 150-fold for acute myeloid leukemia (AML) and 20-fold for acute lymphoblastic leukemia (ALL).¹  Even more striking, young children (<4 years old) with DS have an estimated 500-fold increased incidence of acute megakaryoblastic leukemia (AMKL, FAB subtype M7), now defined in the World Health Organization classification as myeloid leukemia associated with Down syndrome (ML-DS).^2,3  ML-DS is further characterized by a hallmark somatic mutation in the essential hematopoietic transcription factor GATA1 that occurs in nearly all cases of ML-DS, but is absent in non-DS childhood AML.⁴  The GATA1 mutation contributes to another distinctive characteristic of ML-DS — an increased susceptibility of ML-DS leukemic blasts to chemotherapy, in particular to cytarabine, due to effects on cytidine deaminase gene expression.^5,6  Other unique clinical characteristics of ML-DS include the high prevalence of antecedent cytopenias, particularly thrombocytopenia, which is often mistaken for idiopathic thrombocytopenic purpura (ITP), the low incidence of central nervous system disease, and the absence of common chromosomal translocations and inversions such as t(8;21) and inv(16) that are associated with non-DS childhood AML. It is the cure rate of ML-DS, however, that is most remarkable as compared to non-DS childhood AML. Children with ML-DS have far superior outcomes as compared to children without DS and AML with a five-year overall survival (OS) of approximately 93 percent with contemporary pediatric chemotherapy protocols, whereas children without DS and AML have an estimated OS of 60 to 70 percent.^7-10 

Historically, the superior outcomes were not appreciated. From the 1960s through the 1980s, children with DS and AML received minimal if any chemotherapy due to the physicians’ impressions that children with DS and AML were unable to tolerate the intensive chemotherapy regimens designed for childhood AML and parents who declined chemotherapy for their child.¹¹  In the mid-1980s, however, retrospective analyses revealed that outcomes for children with DS were as good as those for children without DS and possibly better, even with lower-dose chemotherapy regimens.^12-14  More specifically, it was revealed that children with DS not only had superior outcomes when treated with less intensive chemotherapy regimens, but also suffered excess toxicity when treated with intensive therapy designed for patients without DS and with pediatric AML.^15,16  Accordingly, clinical trial design for children with ML-DS during the past 20 years in the United States, Europe, and Japan has focused on optimizing and fine-tuning DS-specific chemotherapy regimens that reduce the intensity of the chemotherapy and minimize toxicity while maintaining the high cure rates for ML-DS. These trials have demonstrated that the cumulative doses of anthracycline, cytarabine, etoposide, and intrathecal treatments can be reduced without impacting the excellent survival in this vulnerable patient population.^7-10 

In the current report, Dr. Johann Hitzler and colleagues on behalf of the Children's Oncology Group report the results of the most recent trial toward this effort, AAML1531, which implemented the first risk-stratified treatment protocol for patients with ML-DS. The trial design used minimal residual disease (MRD) for risk group assessment and to assign post-induction standard-risk (SR) or high-risk (HR) therapy; patients with an MRD level of less than 0.05 percent following induction I chemotherapy (thioguanine 50 mg/m²/dose twice daily, days 1 to 4; cytarabine 200 mg/m²/day continuous infusion, days 1 to 4; and daunorubicin 20 mg/m²/dose, days 1 to 4; i.e., TAD [6-thioguanine, cytarabine, and daunorubicin]) were assigned to the SR arm (n=114), and patients with MRD level of more than 0.05 percent were assigned to the HR arm (n=26). SR therapy consisted of two more courses of TAD therapy (induction II and III), followed by two identical “intensification” courses of intermediate-dose cytarabine and etoposide. Importantly, the SR protocol therapy in this trial eliminated the high-dose cytarabine (HiDAC) and asparaginase chemotherapy course used in predecessor trial AAML0431 based on the following rationale: 1) The HiDAC therapy accounted for most adverse events in AAML0431, mostly related to myelosuppression-induced infectious complications⁹ ; 2) The Japan Pediatric Leukemia/Lymphoma Study Group have used a treatment regimen for patients with SR ML-DS that does not contain HiDAC and maintained excellent outcomes⁷ ; 3) MRD assessment following induction I therapy in predecessor trial AAML0431 was highly predictive of treatment outcome with five-year disease-free survival for MRD-negative patients of 92.7 percent versus 76.2 percent for MRD-positive patients.⁹  Thus, the goal of this trial was to identify a subset of patients (using MRD assessments) whose therapy could be de-intensified to further reduce the treatment-related mortality for patients with SR DS-ML, while maintaining a high event-free survival (EFS). For HR patients on this trial, the high-dose cytarabine courses were retained in the two intensification blocks. The trial enrolled 201 pediatric patients with DS (>90 days and <4 years old) between November 2015 and March 2020. The study was designed to compare the two-year EFS of SR patients to the two-year EFS of 93.5 percent from the predecessor trial AAML0431. The planned interim analysis was performed after 50 percent of the expected EFS events occurred and revealed that the survival of arm A (SR arm) was inferior to that of AAML0431 with an EFS of 85.6 percent. The difference was statistically significant (p=0.0002), and arm A was closed to accrual. The two-year OS was 91 percent. Twelve patients suffered a relapse, and 11 of the relapse events occurred within the first year of therapy. The cumulative incidence of relapse for SR patients was 10.8 percent at two years. The survival following a relapse event was very poor with an estimated OS of 16.7 percent at one year. Of the 12 relapsed patients, at least six were treated with salvage high-dose cytarabine, but HiDAC was not effective in the relapsed setting. The toxicity of the protocol treatment was similar overall to the predecessor trial AAML0431. There were 11 deaths on this study. Only one death was related to toxicity during the induction I therapy, while all other deaths were related to relapse.

This report included detailed chromosome analysis for SR patients and revealed distinct clonal abnormalities as compared to non-DS AML. The most common abnormalities being trisomy 8 (30%), tetrasomy 21 (16%), gain of 1q (11%), loss of 7p (13%), gain of 11q (11%), and complex karyotype (12%). Complex karyotype was found more frequently among SR patients who relapsed (n=4/11) compared to those who did not (n=9/99): 36 percent versus 9 percent (p=0.0248). The cumulative incidence of relapse was significantly higher for SR patients with a complex karyotype compared to SR patients without a complex karyotype (30.8% vs. 7.5%, respectively, p=0.001). There were no cases of common translocations and inversions such as t(8;21) and inv(16) in this cohort.